Generally speaking freight carriers calculate shipping costs based on package size and weight (i.e., volumetric weight). This helps prevent lightweight packages that require a large amount of space from becoming unprofitable for the freight carriers.
When printing a shipping label for a package, a freight carrier employee is required to enter the package's size and weight into a software application that uses this information to calculate the cost of shipping. Typically, freight carrier employees derive this information through hand measurements (e.g., with a tape measure) and may weigh the package on a scale. Hand measurements are prone to error, particularly when packages have an irregular shape. These errors may lead to dissatisfaction and/or financial loss. For example, when a shipping company determines, after shipping costs are negotiated, that a package is larger and/or heavier than reported by the customer, additional costs may assessed. In addition, retailers that pass the shipping costs along to customers typically assume the extra shipping costs associated with these errors. As a result, automated dimensioning systems have been developed to bring more accuracy to package volume calculation.
One such automated dimensioning system uses a light projector to project a light pattern (e.g., point cloud) onto objects (e.g., packages) within a field of view. A range camera, physically offset from the light projector, creates a range image from the light pattern reflected from the packages. Software running on a computing device compares the light pattern in the range image to some reference (e.g., a reference image taken during calibration). Through this comparison, the dimensions of a package may be derived.
In order to dimension a specified size range (e.g., a range of package sizes) accurately, the dimensioning system may require a user to position (i.e., align) the range sensor into a particular pose (i.e., height and orientation). This positioning typically takes place during the installation of the dimensioning system. During positioning the pose is computed relative to a reference (i.e., ground) plane that is typically defined prior to positioning. The process of selecting the reference plane and positioning the range sensor is not easily handled by a typical user, but poorly installed range sensors may result in dimensioning errors or the inability to dimension. Therefore, a need exists for a method to assist the user with the selection of a reference plane and the positioning of a range sensor to ensure good performance of the dimensioning system.